Gas-intake or blocked suction?

[rocketscientist]

I have an interesting question from a younger engineer. I think I know the answer but I wanted to get some additional input.

We have a reaction vessel where CO2 is evolved and a centrifugal pump circulates liquid through a heat exchanger to keep it cool. The tank is agitated by a high flow impeller flowing clockwise and downward in a baffled tank.

The pump frequently fails from cavitation. One engineer says it's caused by solids buildup in the suction. Another says it's gas in the suction. How could we determine the difference?

 

[ione]

Buildup could increase pressure losses on the suction side to such a degree to promote conditions for cavitation to take place.
This is how I would approach this issue. First of all one should have an idea of the time required for buildup to form. If once removed the buildup and restored the initial working conditions, incipient cavitation or cavitation were detected (for example through an acoustic investigation) in time shorter than that required to buildup formation, then buildup wouldn’t probaly be the culprit.

 

[tr1ntx]

Pressure tap on the suction (if possible) to see what it is.

Or perhaps less trouble, measure the flow to the heat exchanger. If it tails off in an orderly fashion, then plummets during cavitation, that is likely caused by solids buildup. If it stays fairly constant, but is less than it should be, that's gas entrainment.

Well, I should have asked first, what do you mean "fails from cavitation"? Pitted, eroded impeller and/or housing? Bearing failure from vibration? Periodic low flow? Is the solids buildup periodically cleaned off?

Of course you could have both situations, buildup restricting flow and decreasing pressure, thus evolving gas in the flowstream.

 

[rocketscientist]

Actually, the "cavitation" in the pump is a function of the solid addition. Adding too much solids speeds up the release of CO2.

An amp probe on the pump shows that current is inverse of flow (so I assume we have a measured flow but may not trust it). In other words, current draw goes up as flow decreases.

I am wondering if we are seeing a density drop in the water column since the suction is side draw with no vortex box. A vortex box if properly designed by separate liquid from gas --- then, we'd know if there was a solids buildup --- after eliminating the gas.

 

[Artisi]

I would go for gas entraiment. There has been a couple of discussions over the past 3- 4 weeks on this subject - it might be well worth you searching out the info.